Speed vs Capacity: Designing Data Systems for Real-Time Scale

Join Aerospike, Start.io, and Contentsquare for an evening of deep technical talks and peer-level discussion on the real trade-offs behind operating modern real-time systems in production - where tail latency, throughput predictability, scalability, and cost efficiency are constantly in tension.

This event will move past theory and into hard-earned production lessons: scaling beyond RAM constraints, migrating high-throughput workloads from Redis to Aerospike, designing architectures driven by real cloud cost models, and uncovering the non-obvious pitfalls of migrating real-time systems from AWS to GCP. Topics range from VPC architecture and traffic steering differences to NVMe and local SSD behavior, and why familiar operational playbooks often fail across clouds.

This event is designed for engineers and architects already running latency-sensitive, high-scale systems - where milliseconds matter, growth is relentless, and infrastructure decisions directly impact performance, reliability, and cost.

Agenda 17:30 – 18:00 - Welcome drinks & networking 18:00 – 18:10 - Opening Remarks – Speed vs Capacity: Designing Data Systems for Real-Time Scale Oshrat Ben-Avi Zabludovitz, Israel Country Manager, Aerospike 18:10 – 18:30 - Scaling Past RAM: Architecture Lessons from Our Redis-to-Aerospike Migration Ilan Huchansky, Data Platform Team Leader, Start.io As Start.io scaled, our legacy Redis-based architecture for the user-profiles database reached significant operational limits, particularly during high-traffic periods, node failures, and scaling events. To maintain our strict performance SLAs while reducing operational overhead, we embarked on a search for a more robust, cost-effective NoSQL solution.

In this session, we will share the architectural lessons learned while evaluating the database landscape - including our comparisons with alternatives - and our ultimate migration to Aerospike’s Hybrid Memory Architecture. We’ll discuss how we moved past RAM constraints to achieve sub-millisecond performance at massive scale without compromising on availability or data consistency. 18:30 – 18:50 - Boosting Cost Efficiency with Cost-Aware Architecture Doron Hoffman, Chief Architect, Contentsquare In this session, we explore the critical role of cost-aware architecture in achieving optimal cost efficiency for modern software systems. As organizations increasingly rely on cloud services, microservices, and distributed computing, understanding and managing costs becomes paramount. We’ll discuss the shift from traditional architecture to cost-aware design and examine how architectural decisions directly impact operational expenses. Through a real-world example, we’ll share lessons learned from evaluating multiple cloud providers and their services using cost models and how rearchitecting with cost as a first-class concern made a measurable difference. Attendees will gain practical insights into designing cost-efficient systems that balance performance, scalability, and financial constraints, enabling more sustainable growth. 18:50 – 19:10 - The Latency Leap: Hard Lessons in Migrating Real-Time Systems from AWS to GCP David Gerchikov, Senior Software Engineer, Aerospike Moving a stable web application between clouds is challenging; migrating a high-throughput, real-time system is far more delicate. When heartbeats are measured in milliseconds, the “invisible” differences between AWS and GCP—from VPC architecture to NVMe and local SSD behavior—can determine whether a migration succeeds or fails. This session goes beyond basic service mapping to explore the real friction points of moving a real-time engine from AWS’s regional model to Google Cloud’s global fabric. We’ll cover networking and traffic-steering gotchas, why AWS operational playbooks often don’t translate, and how storage lifecycle constraints in GCP Local SSDs force changes in node management to preserve predictable performance and reliability. 19:10 – 20:30 - Drinks & networking